trade credit, linked to a temporal gap between delivery and

An Investigation into the Determinants of Trade Credit Use

by French Small and Medium Enterprises

Ydriss ZIANE*

M.O.D.E.M, University of Paris X-Nanterre

Abstract

The aim of this paper is to analyse the role of trade credit in the financing of French small and medium-sized enterprises (SMEs) using specific data collected in a recent survey on small firms. We provide direct evidence that inter-firm credit offered by SMEs is sensitive to the strategic importance of firm size and position. Using existing theories, we also test a supply function and a model of demand for trade credit. On the one hand, results validate supply-side theories relative to a reduction of asymmetric information between the buyer and the seller. On the other hand, econometric exercises show that both a transaction motive, linked to efficient cash management, and a financing motive are significant determinant of trade credit demand. According the financing motive, trade credit is specifically used for the purpose of short-term financing by firms exposed to information asymmetries and bank credit rationing.

Résumé

Dans cette contribution, nous examinons le rôle du crédit interentreprises dans le financement des PME françaises à l’aide d’une base de données construite spécifiquement à partir d’une enquête réalisée auprès de chefs d’entreprise. L’analyse des comportements, en matière de gestion des créances clients, souligne l’importance stratégique de la taille et de la position des entreprises dans le financement des ventes. En référence aux théories existantes, nous testons également une fonction d’offre et un modèle de demande de crédit interentreprises. D’une part, les résultats valident les motifs d’offre liés à une réduction des asymétries d’information entre les contractants. D’autre part, ils confirment l’existence d’une composante transactionnelle, liée à une gestion efficiente de la trésorerie, et d’une composante financière de la demande de crédit interentreprises. Selon cette dernière, le crédit fournisseur est utilisé prioritairement à des fins de financement du cycle d’exploitation par les entreprises les plus exposées aux asymétries d’information et au rationnement du crédit bancaire.

JEL Classification : G30 ; G32

Keywords : Trade credit ; Small and medium-sized enterprises ; Credit rationing ;

Asymmetric information ; France

* Correspondence : MODEM, University of Paris X-Nanterre, 200, avenue de la République, 92001 Nanterre Cedex, France ; tel : 0033 (0) 603350877 ; fax : 0033 (0) 140977784 ; e-mail : [email protected]

1. Introduction

Trade credit is an important source of short-term finance for firms in several countries and

especially in Europe. Dietsch [1998] reports that the amount of trade debts is directly

comparable to the total of bank short-term financing for the French corporate sector.

However, payment terms involved by trade credit constitute a specific source of concern

because of financial consequences for the firm’s health, particularly for smaller ones.

According to estimates, in Europe, late payments are at the origin of one insolvency case on

four. The recent reduction of payment terms observed in France has mainly consolidated the

benefit position of larger firms, to the detriment of small and medium-sized enterprises for

which high exposition to asymmetric information is a major obstacle to find external

financing. Acting to limit these adverse effects runs up against a lack of information on the

nature and practices of trade credit. In this contribution, we use specific data collected in a

recent survey to provide evidence on trade credit practices for a panel of French small

businesses observed in 2002. This detailed data set allows a descriptive analysis of

quantitative and qualitative aspects of firms’ accounts receivable management. To check the

validity of different theoretical explanations justifying the use of trade credit, we also test a

supply function and a simple partial equilibrium model of trade credit demand. With regards

to supply, we examine how the level of accounts receivable depends on characteristics

reflecting theoretical motives for offering trade credit. With regard to demand, we estimate a

model including a transaction motive, associated to an objective of efficient cash

management, and a financing motive based on the predominance of credit rationing and

information asymmetries to account for trade credit use.

The remainder of the paper is structured as follows. In section II, we discuss the nature and

practices of trade credit in Europe and particularly in France. Section III presents theoretical

explanations for trade credit. In section IV, we describe the data. A statistical analysis of data

is carried out in section V. Section VI examines the determinants of trade credit offered and

received by firms.

2

2. Trade credit practices

The mechanism of trade credit, linked to a temporal gap between delivery and payment of a

product (or a service), is an important and relatively specific form of financing. It represents

an account receivable (for the seller) and an account payable (for the buyer) of very low

maturity and specific nature since it is relative to transactional and not purely financial

considerations. Trade credit terms are not legally fixed1 but fluctuate according to industry

practice and bargaining power of contractors. Trade credit is multifaceted and for the buyer its

form determines financial costs associated to that financing. Various trade credit

arrangements are illustrated below.

Figure 1

Trade credit configurations

Delivery Discount date Net date T0 T1 T2 Time Discount Net Period Period

Prepayment Cash Credit Late

In the case of payment before delivery, the seller benefits from a credit granted by the

purchaser for the duration which separates that advance from delivery. In the case of cash

payment on delivery and without any reduction of the selling price, there is no credit. For

payment after delivery, the seller extends a credit to the buyer for a period freely fixed by

both contractors. In practice, two scenarios are possible. In the case of a transaction without

discount, at net terms, full payment is due at net date (T2) and the credit extended corresponds

to the period between T0 and T2 (generally 30 days End Of Month, i.e. thirty days from the

next month end). In the case of terms including a mechanism of discount with two-part

terms2, i.e. a reduction of the selling price for a fast or immediate payment, the buyer’s

attitude to payment determines the cost of credit. In case of payment during the discount

period (between T0 and T1), the buyer obtains a price discount and trade credit is free.

Conversely, if payment is made after this date, during the net period (between T1 and T2),

1 Except in the case of statutory payment terms like for perishable goods. 2 An example of two-part terms is “1/10 net 30”, meaning the buyer obtain a 1% discount by paying within 10 days otherwise the full payment is required in 30 days after the delivery.

3

full price is due and the buyer undergoes a financial cost3 linked to the opportunity cost of

foregoing the discount. Only trade credit with two-part terms is, in case of renunciation,

source of high financial cost for the buyer. Late payment defines any payment intervening,

with or without discount, after the net date (T2). Predominantly, mechanisms of discount

include in France a reduction from 0.5 to 5% of the selling price for an immediate4 or fast

payment, generally within 10 days. Table 1 presents the interest cost of foregoing the discount

period for these different values of discount rate according to a net period fixed at 30 days.

TABLE 1

On his part the buyer undergoes management costs of trade credit. He assumes, like a

financial intermediary, specific financial functions (credit risk assessment and bearing,

collecting and financing of receivables). This exercise involves important risks5, particularly

in France (Dietsch [1990]), but can be externalized6. Because of asymmetric information and

size effects, that management cost is higher for smaller firms. Table 2 presents estimates of

that cost, as a percentage of turnover, for different European countries.

TABLE 2

Trade credit practices and common payment terms vary appreciably according to financial

systems [table 3]. In Europe7, one traditionally opposes, according to their payment practices,

northern and southern economies of the continent. Thus, Scandinavian countries, the

Netherlands, the United Kingdom and Germany are countries where payment terms are

traditionally the shortest. Conversely, Belgium, France, Spain, Portugal and Italy report the

longest. These differences can be explained by financial but also institutional considerations

(Marotta [2000]). Financially speaking, widespread practice of discount with two-part terms,

in Germany for example, is a factor of negative influence over the terms of payment as well

as the systematic application of pecuniary penalties for late payments. Institutionally

speaking, efficiency of legal systems relative to the protection of creditor’s claims and the

degree of application of specific commercial laws, as the reservation-of-ownership have an

impact, even partial, on payment habits.

TABLE 3

3 The implicit interest rate is : {(100/(100 – % of discount)) 360/(net period – discount period) – 1}. 4 In this particular case, the reasoning is unchanged but the discount period is nil. 5 By example, for an unpaid of 1000 euros, a firm is constrained to realise additional sales for 100 000 euros (with a gross margin profit = 1%), 20 000 euros (= 5%), or 10 000 euros (= 10%). 6 For a detailed approach of management of receivables, see Mian and Smith [1992]. 7 For a recent comparative study of trade credit practice in Europe, see Hol and van der Wijst [2002].

4

At a macroeconomic level, trade credit generates receivables and liabilities whose weight is

preponderant in the majority of developed countries8. Thus, in proportion of GDP for year

1993, these amounts represent from 3,5 % in Germany to 50% in France according to trade

credit practices [table 4]. For companies, data also underline the strategic role of balance

sheets related to trade credit policies9 [table 5]. In East and Central European countries, which

are exposed to a high level of uncertainty10, practices of inter-firm credit are not developed

(Johnson, McMillan and Woodruff [2002]).

TABLE 4

TABLE 5

Hence, trade credit appears as a key element of the firm’s financial environment in Western

European countries. But the lengthening of payment terms and the proportion of late

payments [table 3] result in increasing the firm’s cash requirements. Moreover, the very long

payment terms practiced by public administrations constitute a source of particular concern.

This situation is accompanied, in France and in Italy, by a recent structural reallocation of

trade credit benefits from smaller to larger firms (Dietsch and Kremp [1998], Marotta [2001]).

Because of the vulnerability of their financial situation and because they depend on a limited

number of customers, smaller firms are in their turn constrained to extend payments of their

suppliers. These circumstances result in a degradation of the business climate and in the

multiplication of risks of chains bankruptcy. This recent evolution appears as an aggravating

fact of the structural component of trade credit according to which payment terms tend to be

shortened in times of economic growth and to lengthen in times of recession. To limit late

payments and promote harmonization across countries, European authorities, in a 1995

recommendation11, have proposed principles and methods to reduce payment terms in Europe.

Noting the inefficiency of those first actions since : « The situation indicates that late payment

has become the rule rather than the exception, with a large number of firms exceeding the

contractually agreed payment terms, thereby creating a large volume of overdue outstanding

in Europe12 », the European Commission voted, in 2000, a directive13 intended to ensure the

maintenance of reasonable payment terms in commercial transactions. The 2001 French law,

8 For an analysis concerning developing countries, see Cuevas and al. [1993], Fafchamps [1997] [2000]. 9 For a synthesis of the evolution of payment terms in France since 1992, see Bardes and al. [2002]. 10 Interactions between macroeconomic uncertainty and trade credit are discussed in Baum and al. [2003]. 11 Commission Recommendation of 12 May 1995, O.J.E.C n°L 127 of 10.06.1995, p.19. 12 O.J.E.C (Official Journal of the European Communities) n°C 216 of 17.7.1997, p.10. 13 Commission Directive E.C 2000-35 of 29 June 2000, O.J.E.C n°L 200 of 8.8.2000, p.35.

5

relative to new economic regulations14, transposes those dispositions. A first evaluation of the

effects of this directive is expected in 2004.

3. Theories of trade credit

A large part of research on the matter, relative to macroeconomic considerations, focused on

the effects of trade credit on the efficiency of monetary policies15. They test the existence of a

trade credit channel or « Meltzer effect » [1960] according to which the flexibility of

financing relationships between large and small companies makes it possible to offset the

consequences of a restrictive shock of monetary policy. Firms with direct access to capital

markets, i.e. larger firms, help out small financially constrained firms (Bernanke and Blinder

[1988]) by extending more trade credit when credit rationing is relevant. According to those

theories, those elements partially explain the difficulty of evaluation and checking of the

broad credit channel16. On the microeconomic level, theories justifying the use of trade credit

are based on the imperfect character of product markets by underlining the major role of

information asymmetries. In this section, we summarize the principal characteristics of these

theories depending if they justify supply or demand for trade credit.

3.1. Theories of trade credit supply

Trade credit supply, materialised by the amount of accounts receivable, is justifying by

verification and marketing considerations but also by motives related to the seller compliance,

the reduction of asymmetric information and the achievement of economies of scale.

The verification motive

Asymmetric information is relative to the firm’s financial activities but also to products.

According to the verification motive, trade credit is a means offered to clients to make sure

that quality and quantity of goods delivered by the supplier are conform before payment

(Smith [1987]). It serves as a process of signalling the quality of products or services sold

through the inspection period it admits. This argument is particularly valid if the quality of

delivered products is longer to assess (high technology or new products) or when the client

holds little information on the supplier. Thus, small firms have to finance a great part of their

sales by trade credit because of a lack of reputation (Long and al. [1993]). In the event of

14 For a critical approach of the transposition of that directive in the French law context, see Drault [2001]. 15 See Meltzer [1960], Brechling and Lipsey [1963], Laffer [1970], Jaffee [1971]. 16 For a recent empirical approach, see Kohler and al. [2000], Nilsen [2002] and Mateut and al. [2003].

6

opportunist behaviours relative to a weak quality of products, the use of discounts with two-

part terms can be a means of reduction of the inspection period (Lee and Stowe [1993]).

Moreover, because of the church tower principle17, this check period is supposed to be longer

when contractors are geographically distant, in the case of export for example.

The sales-promotion motive

Trade credit terms offered and the attitude the seller adopts to enforce them are, as part of the

price, elements of differentiation of suppliers. They can be used to reinforce the competitive

position of firms. For the purpose of sales promotion, developing customer relationships or

attenuation of demand variations, an adequate specification of trade credit terms is possible

since suppliers are not constrained, unlike a financial intermediary, to realize a profit on the

financial part of their commercial transactions. Nadiri [1969] notes that trade credit terms can

be adjusted to gain or preserve market shares over competitors or to offload excess

inventories18. Trade credit is also used to facilitate the establishment of long-term trade

relationships with customers, in particular for small and young firms because of asymmetric

information and reputation effects (Summers and Wilson [2002a]). Trade credit is then a

specific short-term investment whose long-term goal is the certainty of a certain level of sales

related to the existence of stable and mature customer relationships19. In the event of seasonal

or highly variable demand, adjusting trade credit terms can offset uncertainty by increasing

sales when demand is low. Trade credit makes it thus possible to stabilize the demand level

by smoothing business or seasonal cycles without price variations (Emery [1987]).

The price-discrimination motive

Trade credit can be used to price discrimination among customers (Schwartz and Whitcomb

[1978], Brennan and al. [1988], Petersen and Rajan [1997]). In that case, terms of credit,

defined according to industry practices and consequently insensitive to the credit quality of

the buyer20, reduces the effective price of products or services for riskier customers by

partially eliminating the risk premium relative to their status of low-quality borrowers. The

supplier’s advantage lies in the expression of a higher total demand. Subsequently, the

magnitude of the profit margin recorded by the supplier is a principal determinant of the

influence of the price-discrimination motive. The higher the supplier’s profit margin, the more 17 This principle is identified in papers relative to bank credit, for details see Carling and Lundberg [2002]. 18 This process notably permits to transfer storage costs to customers. 19 In this situation, trade credit gives the opportunity for the supplier to become an implicit stakeholder interested in future activities of its customer base (Smith [1987]). 20 For a justification of this hypothesis, see Smith [1980] and Petersen and Rajan [1994].

7

likely the firm is to finance sales. Supported by the inelastic demand of risky borrowers,

price-discrimination is carried out to the detriment of high-quality customers. However, the

use of trade credit discounts, source of high financial cost in case of renunciation, is a means

of limiting opportunist behaviours on behalf of risky customers. It also allows other buyers to

benefit from favourable financing conditions within the framework of long-term relationships

for example. This motive can be explained by the fact that the supplier’s income does not

depend on the financial component of sales.

The seller compliance motive

In a context of limited enforceability of debt contracts, trade credit supply is justified by the

comparative advantage of suppliers over banks in lending to their customers (Cunat [2001]).

This advantage is mainly due to the nature of the trade relation of dependence that underlies

credit and to the extra threat of stopping sales of products and services21 in case of repeated

late payments. The influence of this argument varies appreciably according to the position of

the supplier in vertical competition on the one hand and, on the other hand, the bargaining

power of its customers (Dietsch [1998]). To the detriment of smaller firms, that bargaining

power is often expressed through the respective size of contractors. Probability of loss of the

surplus related to the existence and the maintenance of a long-term trade relationship is a

factor determining the higher lending capacity of suppliers. In a preventive way, Smith [1987]

shows that the use of two-part terms discount is an efficient process of credit risk

identification. Thus, information related to the execution of financial commitments by

customers is relevant because of reduced information asymmetries allowed by trade credit.

The specific-investments motive

The sales-promotion motive has highlighted the importance of short-term investment

represented by trade credit in a perspective of anticipation for future sales. These profits are

relative to the exploitation of long-term trade relationships by the supplier. Consequently, the

seller must make this investment profitable by maintaining, with most faithful customers,

trade relations for a sufficient length of time (Smith [1987]). This is made possible by a

specific use of credit terms. For example, a supplier frequently lengthens payment terms22 or

the discount period23 for customers facing temporary difficulties in connection with an

immaterial and specific investment in relationships. In parallel, Ng and al. [1999] stress that 21 For a discussion of this point under the view of contract theory, refer to Fafchamps and al. [1995]. 22 See Summers and Wilson [2002a] [2003]. 23 Ng and al. report that only 31.8% of firms indicate they do not permit customers to take unearned discounts.

8

an industry specific investment by buyers is also a reason for offering trade credit. That

material investment, which is often specific to industry and unrecoverable in case of

bankruptcy, reduces asymmetric information by creating an information/reputation capital.

However, the value of material investments depends on the vertical position of the firm on its

market, supply of trade credit will not be similar to manufacturers, wholesalers or retailers.

The scale economies motive

According to this motive, firm’s size is an aspect of positive influence on the offering of trade

credit (Ng and al. [1999]). Due to fixed costs, associated with managing accounts receivable,

a large number of customers implies a weaker average management cost and favours the

development of trade credit supply by the means of economies of scale. The argument is

similar for the management costs generated by the practice of discount with two-part terms.

That motive supposes the existence of a positive relation, often checked, between the firm

size and the number of its customers.

3.2. Theories of trade credit demand

To explain the amount of firm accounts payable, representative of the quantity of trade credit

demanded, it’s necessary to take into account credit market imperfections caused by

asymmetric information. Indeed, it allows considering a financing motive (or liquidity

motive) in connection with the macroeconomic considerations of the trade credit channel, in

complement of a transaction motive related to an efficient cash management. The financing

motive, linked to the phenomenon of bank credit rationing, is particularly relevant for small

firms. The combination of transaction and financing motives allows, on the basis of work by

Chant and Walker [1988], the development of a partial equilibrium model of trade credit

demand.

The transaction motive

The transaction motive is based on the simplification of cash management that trade credit

allows (Ferris [1981]). The advantage of trade credit lies in a possible separation between

delivery of products or services, sometimes uncertain, and payment of these goods. For the

buyer, this temporal separation implies savings in transaction costs and a better knowledge of

cash needs. Transaction costs are relative to mobilization, detention and transformation into

means of payment of the firm’s liquid assets at ends of settlement of commercial transactions.

9

The financing motive

Information asymmetries on bank credit markets cause credit rationing (Stiglitz and Weiss

[1981]). Firms with high risk undergo the strictest consequences of funds restriction imposed

by financial intermediaries. In that situation, trade credit demand for financing motive is

justified as a means of limiting these adverse effects, most liquid firms (i.e. larger firms)

helping out constrained firms (i.e. smaller firms) to finance their exploitation cycle (Schwartz

[1974])24. This function of « intermediary of substitution » exerted by the supplier is

facilitated by the nature of the trade relation which underlies the financing. Compared to a

classical financial intermediary, this relation makes it possible for the seller to benefit from

savings in collecting information about small firms, evaluating default risk and valuating

products in case of bankruptcy (Petersen and Rajan [1997]). Thus, advantages in acquisition

of better and less costly information, control of the buyer and valorisation of assets account

for higher efficiency of trade credit with respect to bank credit25. These advantages explain

why the supplier is not forced to limit the amount of loans because of adverse selection and

hazard moral problems that do not allow banks to identify small firms’ credit risk efficiently.

Conversely, by paying late or by dropping the discount period, the borrower automatically

reveals his class of risk to the supplier (Elliehausen and Wolken [1993]). Thus, renouncing

the discount is interpreted as a preventive sign of degradation of the customer’s financial

situation but, more generally speaking, observation of payment terms and their evolutions is

also revealing. Within this framework, the quantity of credit obtained from suppliers appears

as a positive sign of the borrower’s repayment capacity. This signal, intended for financial

intermediaries, can be used to obtain more important bank financing (Biais and Gollier

[1997]).

A simple model of trade credit demand

According to the theoretical elements mentioned above, the anticipated demand for trade

credit of the firm (DTC*) results from a transaction component (TC) and a financing

component (FC), as :

DTC* = TC + FC [1]

Following Elliehausen and Wolken [1993]26, the transaction component (TC) for trade credit

demand can be modelled by variables representative of transaction costs linked to commercial

24 See Emery [1984], Smith [1987], Wilner [2000], Cunat [2001] and Jain [2001]. 25 For a comparative approach of trade credit and bank credit, refer to Dietsch [1998]. 26 The model presented by Elliehausen and Wolken [1993] is based on works by Chant and Walker [1988].

10

relations with suppliers : the volume of purchases (PUR), the uncertainty or temporal

variability of deliveries (∆DEL), the return on liquid assets (RLA) and the conversion costs of

liquid assets into means of payment (CCLA), as :

TC = F (PUR, ∆DEL, RLA, CCLA) [2]

with δTC/δPUR, δTC/δ ∆DEL, δTC/δRLA and δTC/δCCLA > 0

In connection with the exposition to bank credit rationing, the financing component (FC) is

specified according to the financial risk (FR) and the business risk (BR) presented by the firm.

Prices of trade credit (PTC) and bank credit (PBC) are also considered. The demand for bank

credit (DBC) is represented as a decreasing linear function of price. So :

FC = F (FR, BR, PTC, PBC) [3]

with δFC/δFR et δFC/δBR > 0, δFC/δPTC < 0, δFC/δPBC < or > 0

and DBC = F (PBC) [4]

with δDCB/δPCB < 0

Assuming DTC* = TC + FC is the anticipated trade credit demand and AP is the amount of

accounts payable, the realized demand for trade credit (DTC) is written as :

DTC = AP [5]

If part of the anticipated demand (DTC*) for trade credit is not satisfied, it’s due to an

important risk presented by the firm, as :

DTC*- DTC = F (FR, BR) [6]

with δ(DTC*- DTC)/δFR et δ(DTC*- DTC)/δBR > 0

combining equations [5] and [6] yields :

DTC* = AP + F (FR, BR) [7]

The reduced and estimable form of the model27 is obtained by introducing equations [4] and

[7] into equation [1] :

DTC = AP = α + TC + FC + ε

DTC = AP = INT + F (PUR, ∆DEL, RLA, CCLA) + F (FR, BR, PTC, DBC) + ε [8]

The realized demand for trade credit appears as a simple estimable equation including an

intercept (α), a transaction component (TC) and a financing component (FC).

27 For more details about algebraic manipulations, see appendix 1.

11

4. Data description

The data used in this study is from French small and medium-sized enterprises (SMEs) that

answered to a postal questionnaire survey carried out in the beginning of 2003 and concerning

inter-firm credit during year 2002. A questionnaire, tested beforehand, was sent to a panel of

4000 SMEs stemming from a database called D.I.A.N.E28. Firms were selected according to

updated information and available financial statements. In accordance with the European

definition29, selected firms were small and medium businesses [appendix 2]. Moreover,

because of specificities in management presented by quoted firms and those belonging to the

sectors of administration and financial activities, those firms were excluded from the analysis.

The number of exploitable answers, after eliminating questionnaires with missing answers

and after checking the coherence, amounts to 513 or 12.8% of the panel. The survey collected

detailed data on customer and supplier relationships, credit policy, credit practices, product

and market characteristics and bank financing. Thereafter, general and accounting information

contained in balance sheets and income statements were extracted from the database and

completed the survey.

The demographic characteristics of the survey, presented in table 6, show that most of the

respondent firms are public companies (244) or proprietorships (130). The various positions

on the chain of production/distribution are, except for merchant dealers (3), significantly

represented. The industry classification, established by the French aggregated economic

classification for activities (N.E.S)30, underlines the prevalence of firms from trade,

manufacture of intermediate goods and construction respectively. The distribution by type of

market is well balanced. The firm’s size, measured by various indicators (employees, type of

firms, total assets and turnover), indicates a predominance of smaller firms for which, in

theory, the impact of asymmetric information is higher.

TABLE 6

28 « DIsc for Economic Analysis » is a database edited by Bureau Van Dijk, Paris. 29 Commission Recommendation (96/280/EC) of 3 April 1996 concerning the definition of small and medium-sized enterprises, O.J n°L107 of 30.04.1996. 30 The French economic summary classification (NES), taken up by INSEE is an aggregated double entry classification - economic activities and products - relevant for economic analysis purpose. For more details, see : http://www.insee.fr/en/nom_def_met/nomenclatures/nes/pages/nes.htm

12

5. Statistical analysis

This section provides a descriptive data analysis of the survey. These evidences31, linked with

the management of accounts receivable, concern the determinants of inter-firm credit terms,

motives for varying credit terms, average credit period, use of discount with two-part terms

and the frequency of late payments.

Determinants of credit terms

According to the total answers obtained [table 7], internal decisions (44%) predominantly

determine credit terms offered by firms. They precede requests emanating from customers

(35%) and, in last position, influence of industry standards (21%). This classification, as

presented in table 8, is related to the firm’s position on the production and commercialisation

chain. Thus, firms being upstream from this chain (manufacturer, producer, contractor,

wholesaler) have a more important power of negotiation. For those firms, that advantage is an

element justifying the determination of credit terms mainly according to internal decisions.

Conversely, firms located downstream from the production and commercialisation process

(retailer and service provider), enjoying a weaker bargaining power and exposed to a higher

competition level, classify requests from customer (50.9% and 40.7% respectively) as the

principal factor influencing choice of credit terms. The firm’s position on the production and

commercialisation process is an element which determines credit terms. Negotiation power is

also positively related to the firm’s size [table 9]. Micro enterprises classify customer’s

requests as the first determinant of credit terms (42.2%), which is not the case of small and

medium-sized enterprises of the panel (32.8% and 30.3%, respectively).

TABLE 7

TABLE 8

TABLE 9

Motives for varying credit terms

The study of motives for varying credit terms offered reveals behaviours strongly depending

on the firm’s size. The classification of ten motives for variation on a Likert five-point scale32

underlines the respective importance of financing and competitive motives, contrary to sales

promotion motives, in the modification of credit terms offered by the SMEs [table 10].

31 In this paper, we only present a part of survey evidences but all evidences are available from author. 32 1= not at all important, 2= unimportant, 3= neither important nor unimportant, 4= important, 5= very important.

13

TABLE 10

« To help a customer in temporary difficulties » appears as the principal reason for varying

credit terms. Consequently, the financing or liquidity motive and the role of « intermediary of

substitution » exerted by suppliers are checked empirically. However, size classification

shows that financial aid is stronger as the firm’s size is reduced33 [tables 11A-11B-11C]. The

largest firms, i.e. most liquid firms, are not those that ensure the most significant support in

the event of customers’ financial difficulties. Conversely, medium firms, compared to small

and micro-sized firms, attach more importance to motives which allow to take advantage of

using trade credit34 (« to improve cash flow» and « in response to economic conditions »).

TABLE 11A

TABLE 11B

TABLE 11C

In relation with competitive motives, « to retain a customer » is the major reason for varying

credit terms for all size class. However, it seems that, in connection with bargaining power

and in a counterintuitive way, micro-sized firms attach little interest to motives « because a

customer insists » (4.8% of micro firms consider this motive as very important against 8.4%

and 8.7% of small and medium-sized firms) to justify a modification of credit terms. It is the

same, to a lesser extent, with the motive of « to remain competitive ». Modification of trade

credit terms for competitive reasons appears all the more realizable or desirable that a firm’s

size and sales are significant.

Duration of credit period

Answers concerning the duration of inter-firm credit [table 12] stresses that, for a majority of

firms (60.7%), payments do not occur before 47 days. The most frequent payment period is

“60 days and more” (34.4%) whereas 5.4% of firms are generally paid cash. Comparatively35,

it is interesting to note that, while the percentage of firms paying cash seems stable in

developed countries, approximately 5% (4.8% in United Kingdom for Summers and Wilson

[2002a], 4.2% in United States for Ng and al. [1999]), the most frequent payment period in

France (60 days and more) is double than that observed in U.K (30 days for Summers and

33 18.3% of micro-enterprises declare that the motive « to help a customer in temporary difficulties » is very important, 15.8% of small enterprises and 13.6% of medium enterprises, tables 11A-11B-11C respectively. 34 15.4% and 9.2% of medium enterprises declare that the motives « to improve cash flow » and « in response to economic conditions » are very important versus 11.7% and 7.8% of small enterprises, 6.8% and 5.4% of micro-enterprises, tables 11A-11B-11C. These results validate Dietsch and Kremp [1998]. 35 The comparisons are made only with papers adopting the same reasoning, that is to say giving empirical results from recent self-made surveys : 500 and 655 firms in Summers and Wilson [2002a] and [2003], 950 firms in Ng, Smith and Smith [1999] and 1549 firms in Marotta [2001].

14

Wilson [2002a] and [2003]) and in U.S (30 days for Ng and al. [1999]), but is only half the

average credit period in Italy (124 days for Marotta [2001]).

TABLE 12

The firm’s market size is a major determinant of credit periods [table 13]. Most than half of

firms (52.1%) working on an international sales market have an average credit period of “60

days and more” against 17.6% for firms with local sales market. Those elements confirm the

validity of offering trade credit for a checking motive. In relation with the scale economies

motive, the firm’s size positively influences the average credit period [table 14].

TABLE 13

TABLE 14

The use of the two-part terms discount in credit offered

There is little evidence on the use of the discount mechanism by French small and medium-

sized businesses36. Nevertheless, that information determines the financial cost of trade credit

for buyers. The proportion of firms offering frequently or always is 10%. Terms of discount

generally involve a price reduction of 1 to 3% for a fast payment [table 15]. Half of firms

(49.3%) never use it, 25% seldom do and 15.6% occasionally do. The practice of offering

two-part terms discount is mostly developed in Anglo-Saxon countries (17% and 20% of

SMEs in U.K37, 37% and 24% of SMEs38 and 25.3% of large enterprises39 in U.S) but

comparable in Italy (8.2% of firms for Marotta [2001]). Our data confirms that offering

discount is strongly dependent on industry practices (Ng and al. [1999]) and is primarily

concentrated in sectors of manufacture of consumer goods, intermediate goods and trade.

Moreover, firms with international sales, exposed to asymmetric information about buyers,

use the process of discount for early payment more intensively [table 17].

TABLE 15

TABLE 16

TABLE 17

36 A sample survey by Louis Harris in 2001 indicates that 85% of French SMEs never use trade credit discounts with two-part terms. 37 Summers and Wilson [2002a] and [2002b], respectively. 38 Elliehausen and Wolken [1993] and Danielson and Scott [2000], respectively. 39 Ng, Smith and Smith [1999].

15

The frequency of late payments

Paying late is a widespread practice for the firm’s customers of the survey [table 18]. Only

23.4% of firms declare never (4.1%) or seldom (19.3%) undergo late payments in the

management of accounts receivable. For 41.7% of firms, paying late is occasional but

frequent in 32.6% of cases and systematic for 2.3% of the sample.

TABLE 18

This report confirms the apprehension of European authorities to see late payments become

the rule rather than the exception. This situation is, in the French case, worsened by the bad

payment habit of public administrations as underlined by the non-isolated remark of this

manager : « If the State and public firms initially respected legal payment terms, that would

give the example, public administration is the sector with the longest payment terms in spite

of legislations and customer relationships, 120 days on average instead of 30 to 45 days

expected ! »40. Charging systematic penalties in the event of late payment is envisaged by new

regulations but the application to small and medium-sized enterprises seems problematic as

the following comments attests: « It is impossible, commercially, to invoice financial interests

to customers [...] »41; « The attitude consisting in refusing customers who practise late

payments can be considered as heroic so much as the principle of late payment is firmly

anchored in French mentalities and the majority of suppliers are ready to accept anything in

order to sell [...], up to now new measures taken by public authorities have remained dead

letter. »42

40 Comment remark made by the manager of a small enterprise of general electricity created in 1997. 41 Comment remark made by the manager of a medium enterprise of clothing sales created in 1993. 42 Comment remark made by the manager of a medium enterprise of printing created in 1987.

16

6. Econometric investigations

The descriptive analysis stressed the importance of the firm’s characteristics in the

determination of trade credit terms. To look further into these relations and to test the exposed

theories, this section presents the results of econometric analysis of both, the supply of trade

credit offered by firms and the model of trade credit demand derived from work by

Elliehausen and Wolken [1993].

6.1. Determinants of trade credit supply

In this first empirical approach, we mainly use information contained in the survey to estimate

the determinants of accounts receivable, the duration of credit and trade credit terms choices.

We test how these dependent variables, representative of trade credit supply, depend on

various characteristics relative to the above theories.

6.1.1. The dependent variables

Trade credit importance and duration - The quantitative aspect of trade credit supply is

approximated by the firm’s accounts receivable to annual turnover ratio (TCI). This variable is

estimated by the ordinary least square method. Trade credit duration (TCD), expressed in days

of turnover and classified in an ordered scale, is modelled by an ordered probit procedure.

Trade credit terms - We model the choice of trade credit terms, as in Ng and al. [1999], using

logit estimates to identify the firm’s characteristics according to the most commonly payment

terms proposed to customers43. We construct three dichotomous dependent variables :

offering net terms versus cash terms (NT vs CT), two-part terms versus cash terms (TPT vs CT)

and two-part terms versus net terms (TPT vs NT).

6.1.2 The independent variables

The independent variables, drawn from the theoretical discussion, relate to supplier’s

characteristics offering credit, product characteristics and the nature of demand emanating

from customers.

43 Following survey answers to the question asking the firm’s most commonly used payment terms in financing sales, 437 firms prefer net terms, 48 firms prefer two-part terms and 28 firms prefer cash terms.

17

Supplier’s characteristics

Size – The firm’s size is measured by the logarithm of total assets (LTA). We also consider,

including dummies variables (LOC, REG, NAT, INT)44, the influence of the firm’s market size

(local, regional, national or international) on trade credit supply.

Age – To proxy development, reputation and credit worthiness, the firm’s lifespan (AGE) is

measured according to a simple logarithmic specification [Ln (1 + age of the firm)]. We also

use the square of this expression (AGE²) in order to take into account the nonlinear influence

of age on the quantity of trade credit offered (Petersen and Rajan [1997]).

Profitability - The rate of gross margin of exploitation (PROF) is used as an indicator of profit

with regard to turnover.

Credit rationing - Credit rationing is directly estimated according to the opinion of firm’s

head. The variable (RAT) identifies firms constrained in obtaining short-term bank financing.

Product characteristics

Verification - To test the verification motive, we use the variable (QUA) to identify firms for

which the quality of products or services sold cannot easily be checked in terms of cost and

time. Similarly, fast changing products and innovating products are viewed as characteristics

justifying a longer inspection period for customers (Ng and al. [1999]). These characteristics

are included by dummies variables (FCP) and (INN) respectively.

Competition - In relation with strategic considerations of trade credit supply, the variable

(COM) identifies firms evolving on strongly competitive market.

Demand characteristics

Customer base - To be informed about the frequency of transactions and the possibility of

making economies of scale in the management of accounts receivable, we measure the

number of active customers of the firm (NAC). In addition, we approximate firm sales

concentration by the variable (SCN) defined as annual turnover to number of active customers

ratio. The amount and the form of trade credit also depend on customer type. With dummies

44 When we use different dummies variables to represent a same characteristic, we omit, in regressions, the first dummy variable for reference. Hence, results are interpreted in reference to this omitted variable.

18

variables (RET, WHL, PRO)44, we identify the firm’s most common type of customer (retailer,

wholesaler or producer). Considering the importance of the negotiation power to determine

credit terms, we can measure the relative size of the firm’s most common customers (lower,

equal or higher) by including specific variables (LOW, EQU, HIG)44.

Seasonal variation – Firms which operate in seasonal markets might be expected to use trade

credit as a tool to manage demand variations. We include a dummy variable (SEA) to

distinguish firms facing a seasonal demand from others.

Table 19 provides descriptive statistics for these dependent and independent variables.

TABLE 19

6.1.3. Econometric results

Trade credit importance and duration - Results relating to the determinants of variables (TCI)

and (TCD) are presented, respectively, in columns I-II and III-IV of table 20.

TABLE 20

The firm’s size and age have a positive and significant influence on the amount and the

duration of trade credit. Larger firms have a higher accounts receivable to sales ratio. Thus,

the magnitude of the coefficient of size (LTA) indicates that a firm with 480.000 euros in total

assets extends an additional 2.4% more of its sales in the form of accounts receivable

compared to a firm with 75.000 euros in total assets45. The effect is smaller than in Petersen

and Rajan [1997]. This result confirms the predictions of the scale economies motive for

offering trade credit. Similarly, firm lifespan (AGE) positively influences trade credit supply.

However, the significant coefficients of the quadratic form of the age variable (AGE²) show

that such influence is not linear but only positive during first years of a firm’s life. This effect

suggests a trade credit use related to the acquisition of an efficient firm size particularly

through the sales promotion motive. On our panel, this result is valid for estimates of trade

credit importance (Petersen and Rajan [1997]) and duration (Summers and Wilson [2002a]).

The firm’s market size (REG-NAT-INT) also has a positive impact on the importance and

duration of offered trade credit. In reference to the verification motive, information

asymmetries related to geographical distance between contractors justify this fact. The

positive and significant coefficients of the profitability variable (PROF) highlight the

45 These percentages are calculated in reference to the value of quartiles (1 and 3) of the total assets distribution.

19

relevance of the price-discrimination motive : firms with higher gross profit margin have

more incentive to sell additional units and, consequently, finance a larger part of their sales by

inter-firm credit. Conversely, firms which do not obtain total desired bank-financing (RAT)

offer a reduced and shorter amount of trade credit compared to firms which are not exposed to

bank credit rationing. In reference to the financing or liquidity motive, this suggests trade

credit terms are also determined by the seller’s financial situation.

As far as product characteristics are concerned, the fact that product quality is difficult to

determine by inspection (QUA) because of information asymmetries and necessary inspection

period prior to payment seems to have a positive influence on trade credit offered to

customers. From the supplier’s point of view, a fast changing product limits both, the period

of sale at full price and the collateral value in case of complaint (Summers and Wilson

[2003]). These elements can account for the negative influence of the variable (FCP) over the

amount and the duration of the trade credit offered. The coefficients of variables measuring

the degree of product innovation (INN) and market competition (COM) are statistically

insignificant.

Following the scale economies motive, firms dealing with a high number of active customers

should offer more trade credit than others. These predictions are not checked empirically

since the significant variable (NAC) indicates that a high number of customers is negatively

correlated with the magnitude of accounts receivable and trade credit duration. Management

costs of outstanding debt, supposed higher for smaller firms, can account for that result. For

small and medium-sized enterprises of the panel, the achievement of economies of scale is not

related to the quantitative aspect of the customer base but contrary to the degree of sales

concentration. Working with a reduced number of customers, involving a higher amount of

sales per customer, results in a higher engagement in trade credit as underlined by the variable

(SCN). Considering the bargaining power of customers representing a significant share of the

supplier’s annual turnover, this situation can be as undergone as desired46. Compared to firms

whose customer base is mainly composed of retailers, firms in relation with producers (PRO)

offer a significantly more important and longer credit. On the one hand, this result is

accounted for by the reduction of information asymmetries related to larger specific

46 It appears difficult to separate from an large customer because of a conflict dependent on trade credit terms (undergone situation), but table 10 stresses that 26.9% of firms judge the motive « to attract a new large customer » like important (16.7%) or very important (10.2%) in the decision of varying credit terms (desired situation).

20

investments realized by this type of customer. On the other hand, Ng and al. [1999] stress that

producers face a higher need for financing because of a long production cycle. Indicators of

seasonal demand (SEA) and relative size of the firm’s customers (EQU-HIG) are not

significant.

Trade credit terms - Columns V to VII of table 20 present results of estimates concerning

binary choice dependent variables : net terms versus cash terms (NT vs CT), two-part terms

versus cash terms (TPT vs CT) and two-part terms versus net terms (TPT vs NT).

TABLE 20

Results confirm that firms selling to distant customers extend, in relation with information

asymmetric, more two-part terms in the financing of commercial transactions. In reference to

the seller compliance motive, this discount process allows a better recovery of payment.

Conversely, younger firms (AGE), exposed to a lack of reputation on their market and unable

to justify specific knowledge in management of accounts receivable, are more likely to offer

net terms credit. Net terms provide a classical inspection period for customers to check

quality of products.

In relation with product characteristics and for the above-mentioned reasons (lower collateral

value and reduced period of sale), firms with fast changing products demonstrate a clear

preference for cash terms. Firms selling innovating products (INN) are more likely to offer

net terms versus cash terms. In that case, net terms give the buyer a better opportunity to test

the new product before paying and permit the seller to promote sales. Similarly to Summers

and Wilson [2002a], we find that offering two-part terms is associated with more competitive

markets (COM). This result gives support to the price-discrimination motive because two-part

terms are an efficient process of supply differentiation.

As regard to demand characteristics, small firms in relation with producers (PRO),

comparatively to others, tend to privilege net terms versus cash or two-part terms. Indeed,

financing sales by two-part terms in this case is neither interesting for the seller (producers

make larger specific investments reducing asymmetric information) nor for the buyer

(because of an important need for financing). Other independent variables (NAC, EQU, HIG

and SEA) do not have any significant impact on the choice of trade credit terms.

21

6.2. Determinants of trade credit demand

The aim of this second empirical approach is to test a demand function for trade credit.

Following the model presented above [point 2.2. and appendix 1], use of trade credit for

purchases is explained by a transaction motive, linked with a simplification of cash

management, and a financing motive. According to the latter, because of information

asymmetries, financially constrained firms use trade credit to finance their exploitation cycle.

6.2.1. The model specification

According to equation [8], the reduced and estimable form of the model includes a transaction

and a financing component. The transaction component is modelled as a function of

transaction costs whereas financial risk and exposition to bank credit rationing determine the

financing risk. The explained variable is the trade credit demand realized by firms, following

equation [5] it corresponds to the accounts payable to total assets ratio (AP). To look further

into the financing motive, we can also estimate, using a logit model, the probability that firms

make some of their payments after the due date, using late payments. The dummy dependent

variable (LTP) identifies firms for which late payment is a current practice for financing

purchases47.

The transaction component

As in equation [2], it is necessary to quantify the volume of purchases, the uncertainty or

temporal variability of deliveries but also the return and conversion costs of liquid assets into

means of payment to evaluate savings generated by the use of trade credit in payment of firm

purchases. Relatively to transaction costs, the ratio of stocks to total assets (STO) is used to

calculate the volume of transactions with suppliers. Stocks turnover (STUR), expressed in

days of turnover, and the number of active suppliers (NAS) in relation with the firm indicate

variability or uncertainty in transactions with suppliers. We suppose that firms with high

stocks turnover and trading with numerous suppliers face increased uncertainty and higher

transaction costs. Return on liquid assets and conversion costs are relative to firm’s size.

Indeed, larger firms are more likely to invest their treasury funds on short-term markets. The

volume of annual turnover (TNV) indicates the size of the firm.

47 To the question asking the firm’s frequency use of late payments on purchases to suppliers, 90 firms (17.5%) declare to frequently (69) or always (21) use it.

22

The financing component

Demand for trade credit attributable to the financing motive is linked with financial and

business risk, price of trade credit and price of bank credit (equation [3]). We use a classical

debt ratio defined as the value of total debt to equity (LEV) to approximate financial risk.

Furthermore, we form a financial appreciation of firms using a scoring variable (SCOR). This

financial score is built from the function48 defined by Conan and Holder [1979]. If we

consider the high mortality rate, about 50%, of French small and medium-sized businesses

during their first five year-life49, the business risk can be modelled by the firm lifespan (AGE).

We also differentiate firms for which the head responsibility is total and unlimited in the

event of bankruptcy (Proprietorship and Partnership) from those for which that responsibility

is limited to financial contributions (Limited liabilities proprietorship, Limited liabilities

company, Public companies). Such distinction, established by the dummy variable (RAM),

informs on the degree of risk aversion of managers and completes the modelling of business

risk. As in Elliehausen and Wolken [1993], we model the price trade credit using a cash

discount dummy variable. This variable (TPT) indicates whether suppliers offer the firm two-

part terms discounts for early payments of purchases or not. We know that only trade credit

with discount for early payment implies a financial cost for the buyer [table 1]. The price of

bank credit is approximated, as specified in equation [4], by the quantity of bank credit

demand recorded by the firm. This demand corresponds to the ratio of short-term credit to

total assets (STD). In addition, to investigate if unavailability of credit from financial

institutions affects the financing demand for trade credit, we identify firms exposed to short-

term bank credit rationing by the dummy variable (RAT).

Descriptive statistics relative to transactional and financial variables are presented in table 21.

TABLE 21

6.2.2. Econometric results

Ordinary least squares estimates of the accounts payable variable (AP) are reported in

columns I and II, table 22. This table also reports, in columns III and IV, logit estimates of

probability that firms frequently or systematically pay commercial debts late (LTP).

TABLE 22

48 For details about the function calculation mode : http://diane.bvdep.com/Diane/help/HelpDiane/diadoc10.htm 49 Following the communication of the minister for Small and Medium Enterprises, Trade and Cottage Industry in council of ministers of 3 November 1996 on enterprise creation. For details, see http://www.senat.fr/rap/r96-374/r96-37410.html

23

To a significant extend, the transaction component of the trade credit demand is significant to

explain the amount of accounts payable of firms. The positive and significant coefficient for

the variable (STO) indicates that a higher volume of transactions results in a higher demand

for trade credit. As for temporal variability of transactions, firms in relation with numerous

suppliers (NAS) and those with higher stocks turnover (STUR) finance a more important share

of their purchases by trade credit use. Relatively to return on liquid assets and brokerage

costs, the size variable (TNV) positively affects the demand for trade credit. Consequently,

results in connection with the transaction motive fully validate the expression of a demand for

trade credit linked with efficient cash management.

The set of financing variables also accounts for a significant part of the demand for trade

credit. The scoring variable (SCOR) and the ratio of debt to equity (LEV) reflecting the

financial risk are both significant. This result indicates that firms with high financial risk

make a greater percentage of their purchases using trade credit. It is confirmed by the credit-

rationing variable (RAT). This positive and significant variable demonstrates that

unavailability of credit from financial intermediaries positively affects the demand for trade

credit, suggesting that additional suppliers financing is used to finance the exploitation cycle

of constrained firms. Furthermore, these evidences confirm theories relative to the

informational advantage of suppliers over traditional financial institutions in providing

financing to small firms exposed to asymmetric information. This advantage consists in using

cheaper and more efficient information about clients to monitor accounts receivable (Petersen

and Rajan [1997]).

Only one of the two variables proxying business risk, age of firms (AGE), is significant to

account for the demand for trade credit. Younger firms, with less business experience and

reputation, have greater financing demand for trade credit (Singleton and al. [1999],

Elliehausen and Wolken [1993]). The risk aversion of a firm’s head (RAM), in connection

with the legal form of business, is not a significant variable of the model. Similarly, the

variable reflecting the price of trade credit (TPT), in spite of a negative sign, do not account

for the demand for trade credit. A possible explanation is that trade credit discount with two-

part terms is not widely practiced for firms quoted in the panel. Conversely, the price variable

of bank credit, quantified by the short-term debt to assets ratio (STD), has a positive and

significant effect on demand. That positive sign implies that small and medium-sized firms do

not use short-term bank financing as a substitute but as a complement for trade credit finance.

24

That result, similar to Elliehausen and Wolken [1993], invalidates empirical predictions from

the pecking order theory50 according to which trade credit financing comes lower down than

borrowing from financial intermediaries (Isaksson [2002], Petersen and Rajan [1997]).

The probability that firms practise late payments when purchasing is a means to check the use

of trade credit for a financing motive because firms using trade credit only to simplify their

cash management or to conform to industry practices would not be exposed to adverse effects

of paying suppliers late. According to logit estimates of the dependent variable (LTP),

transaction variables are not, as supposed, significant determinants of that probability (STO,

STUR, NAS, TNV). In contrast, financing variables are statistically significant determinants.

Less experienced (AGE) and strongly indebted firms (STD), like those presenting greater

financial risk (SCOR), tend to pay late frequently or systematically. The positive and

significant variable (RAT) indicates that when firms are financially troubled, i.e. exposed to

bank credit rationing, they try to satisfy their demand for financing by using late payments.

7. Conclusion

By exploiting a survey carried out in 2003 in relation with 513 French small and medium-size

enterprises, a first contribution of this paper is to provide new empirical evidence on the use

of inter-firm credit in the achievement of commercial transactions. Such evidence related to

the determination of trade credit terms (influencing factors, motives for variation and credit

periods) and to common uses (practice of discounts, frequency of late payments), allows a

better understanding of head behaviours as regards to accounts receivable management. They

underline, in particular, that trade credit terms offered to finance sales are sensitive to the

strategic influence of supplier’s size and position. The most frequent credit period observed,

60 days and more, supports the argument that France is a financial system where payment

terms are longest. That situation, dependent on financial (low practice of discounts and

penalties for late payments) and institutional considerations (right of creditors, reservation-of-

ownership), can also be explained by conventions, rules of thumb and trade traditions. On the

European level, heterogeneous practices and situations appear as an important obstacle to the

harmonization promoted by supranational authorities.

50 The pecking order theory is presented in Myers [1984].

25

By using qualitative and quantitative information contained in the survey, a second

contribution is to present results from econometric tests of both empirical determinants of

supply and demand for trade credit. In relation with trade credit supply, the firm’s accounts

receivable is mainly influenced by motives allowing a reduction of asymmetric information

between suppliers and buyers. Larger, most experienced firms and those with higher gross

profit margins expend more trade credit for financing sales. These results highlight the

validity of motives in connection with demand stimulation and achievement of economies of

scale. Firms with a widespread sales market offer a longer and higher trade credit financing

justifying the verification motive. As for demand characteristics, a concentred customer base

implies higher sales financing by suppliers. The choice of trade credit terms seems to be

predominantly influenced by product and market characteristics. With reference to the simple

demand model, estimates show that a transaction component, in touch with an efficient cash

management, and a financing component explain small firms’ demand for trade credit. A

higher volume of purchases and an elevated temporal variability of transactions positively

influence the transactional demand. In accordance with the financing motive, firms most

exposed to information asymmetries manage accounts payable and late payments on purpose

of financing cash needs. It is the case of SMEs presenting poor experience, higher financial

risk and exposed to credit rationing from financial institutions. The complementarity of small

firms’ external financing, trade credit and short-term bank credit, highlights the relevance of

the credit-rationing hypothesis and accounts for the informational advantage of suppliers.

26

Table 1 Opportunity costs of foregoing the discount period

Two-part terms

(% of discount/discount period, net period) Implicit interest cost of foregoing

(in %) 0.5%/10, net 30 1%/10, net 30 2%/10, net 30 3%/10, net 30 4%/10, net 30 5%/10, net 30

9.5 19.8 43.9 73

108.5 151.7

Table 2 Management costs of trade credit in Europe

(means, in % of turnover)

Country Management costs Germany Belgium

Spain France Italy

Netherlands Portugal

United Kingdom

1.1 0.7 1

1.1 1.2 0.8 1.9 0.8

Source : Baromètre Eurofactor (2002)

Table 3 Credit period, debt period and late payments in Europe

(means, in days of turnover)

Source Country Credit period Debt period Late payments

Baromètre Eurofactor

(2002)

Germany Belgium

Spain France Italy

Netherlands Portugal

United Kingdom

44 57 85 68 87 46 63 44

- - - - - - - -

16 10 7

15 15 15 21 10

Dun & Bradstreet

(2001/2002)

Germany Belgium France Italy

Netherlands United Kingdom

30 à 60 45 à 90 60 à 90

60 à 120 25 à 40 30 à 60

- - - - - -

8.9/8.7 18.4/13.9 16/16.5

16.6/16.1 17.7/16.9 14.2/18.2

U.F.B Locabail (1999)

Germany France Italy

United Kingdom G 4

33 65 90 52 61

22 53 71 37 46

8 15 17 19 16

27

Table 4 Importance of trade credit

(means, in % of GDP)

Trade credit Trade debt Country 1983 1993 1983 1993 Germany Canada Spain

United States France Japan

United Kingdom

7.5 20.4 40.1 17.5 42.4 64.4 19.4

5.7 19.3 32.1 15.8 49.3 50.8 14.7

4.6 22.5 30.6 14.2 40.2 52.1 20.3

3.5 21.1 20.8 12.8 39.4 39.0 15.5

Source : Kneeshaw (1995), Table 6

Table 5 Importance of accounts receivable and accounts payable

(means, in % of total assets, total debt and sales)

Source Country A.R/T.A1 A.P/T.A1 A.P/T.D1 A.R/Sales1 A.P/Sales1

Hol, van der Wijst

(2002)

Germany Austria Belgium Denmark

Spain France Italy

Netherlands Portugal Sweden

11 13

17.5 15 26 24 33 6.5 22 7

- - - - - - - - - -

20 15.5 27

17.5 41.5 33.5 37.5 12 28 11

- - - - - - - - - -

- - - - - - - - - -

Bardes (2002)

Germany Spain

United States France Italy Japan

8 25 12 26 33 16

6 23 7

23 24 16

- - - - - -

- - - - - -

- - - - - -

Johnson, Mc Millan, Woodruff

(2002)

Poland Russia

Slovaquia Ukraine

- - - -

- - - -

- - - -

2.7 0

3.4 0.7

- - - -

Petersen, Rajan (1994)

United States Small firms Large firms

- -

- -

- -

4.4

11.6

1.8 7.6

1 A.R=accounts receivable, A.P=accounts payable, TA=total assets, TD=total debt

28

Table 6 Demographic characteristics of survey respondents in 2002

Form of business N % Position N %Proprietorship 130 25.3 Manufacturer 175 34.1 Limited liabilities proprietorship 4 0.8 Producer 11 2.1 Partnership 6 1.2 Contractor 67 13 Limited liabilities company 129 25.1 Wholesaler 113 22 Public limited company 244 47.6 Merchant dealer 3 0.6 Retailer 53 10.3 Service dealer 91 17.9 Industry classification (N.E.S) N % Type of market N % Agriculture, forestry, fishing (EA) 4 0.8 Local 74 14.4 Manufac. of food prod./beverages/tobacco (EB) 16 3.2 Regional 230 44.8 Manufacture of consumers goods (EC) 37 7.3 National 136 26.5 Manufacture of motor vehicles (ED) 8 1.5 International 73 14.3 Manufacture of capital goods (EE)) 32 6.3 Manufacture of intermediate goods (EF) 91 17.7 Construction (EH) 67 13 Age N %Trade (EJ) 185 36 [0-4] 141 27.5 Transports (EK) 28 5.4 [5-9] 194 27.8 Services to businesses (EN) 38 7.4 [10-14] 105 30.4 Personal and domestic services (EP) 6 1.2 [15-19] 51 10 Education, health and social work (EQ) 1 0.2 [20 et +] 22 4.3 Employees N % Type of firms N %[0-9] 147 28.6 Micro-enterprise 147 28.7 [10-19] 98 19.1 Small enterprise 204 39.7 [20-49] 162 31.6 Medium enterprise 162 31.6 [50-99] 61 11.9 [100-250] 45 8.8 Total assets (€000) N % Annual turnover (€000) N %[0-49] 81 15.8 [0-49] 43 8.4 [50-99] 95 18.5 [50-99] 53 10.3 [100-199] 98 19.2 [100-199] 93 18.1 [200-499] 130 25.3 [200-499] 141 27.5 [500-749] 37 7.2 [500-999] 88 17.2 [750-999] 19 3.7 [1000-1999] 48 9.3 [1000 et +] 53 10.3 [2000 et +] 47 9.2

29

30

Table 7

Determinants of credit terms offered

Type of determinants N % firms Industry standards Internal decisions

Requests from customers

109 226 178

21.2 44.1 34.7

Total 513 100

Table 8 Determinants of credit terms offered by position

Type of determinants Manufacturer

N % Producer

N % Contractor

N % Wholesaler

N % Merch. dealer

N % Retailer

N % Service dealer

N % Industry standards Internal decisions


42 73 60

24 41.7 34.3

4 5 2

36.4 45.4 18.2

12 29 26

17.9 43.3 38.8

29 59 25

25.7 52.2 22.1

1 1 1

33.3 33.3 33.3

8 18 27

15.1 34

50.9

13 41 37

14.3 45

40.7 Total 175 100 11 100 67 100 113 100 3 100 53 100 91 100

Table 9 Determinants of credit terms offered by type of firms

Type of determinants Micro-enterprises

N % Small enterprises

N % Medium enterprises

N % Industry standards Internal decisions


25 60 62

17 40.8 42.2

42 95 67

20.6 46.6 32.8

42 71 49

25.9 43.8 30.3

Total 147 100 204 100 162 100

31

Table 10 Motives for varying credit terms offered

Motives for varying credit terms offered [1]

N % [2]

N % [3]

N % [4]

N % [5]

N % Total

N % Sales-promotion motives

To promote existing products/services To promote new products/services

Competitive motives

To attract a new large customer To attract new customers generally

To retain a customer Because a customer insists

To remain competitive

Financing motives To improve cash flow

To help a customer in temporary difficulties In response to economic conditions

237 293

163 195 102 92

150

186 79

131

46.2 57.1

31.8 38

19.9 17.9 29.2

36.3 15.4 25.5

135 85

62 132 75 88 147

120 59 75

26.3 16.6

12.1 25.7 14.6 17.1 28.7

23.3 11.5 14.6

82 73

150 102 160 193 113

83 156 181

16

14.2

29.2 19.9 31.2 37.6 22

16.2 30.4 35.4

38 40

86 51

111 102 64

65 138 87

7.4 7.8

16.7 9.9

21.6 19.9 12.5

12.7 26.9 16.9

21 22

52 33 65 38 39

59 81 39

4.1 4.3

10.2 6.5

12.7 7.5 7.6

11.5 15.8 7.6

513 513

513 513 513 513 513

513 513 513

100 100

100 100 100 100 100

100 100 100

[1]= not at all important, [2]= unimportant, [3]= neither important nor unimportant, [4]= important, [5]= very important

Table 11A Motives for varying credit terms offered by type of firms : the micro-enterprises


N % [2]

N % [3]

N % [4]

N % [5]

N % Total



Competitive motives





To help a customer in temporary difficultiesIn response to economic conditions

64 84

49 55 30 37 43

66 22 42

43.5 57.1

33.3 37.4 20.4 25.2 29.3

44.9 15

28.6

38 27

14 45 24 23 45

35 17 23

25.9 18.4

9.5 30.6 16.3 15.6 30.6

23.8 11.6 15.6

25 17

42 23 41 51 33

23 42 52

17

11.6

28.6 15.6 27.9 34.7 22.4

15.6 28.6 35.4

14 14

28 15 35 29 17

13 39 22

9.5 9.5

19.1 10.2 23.8 19.7 11.6

8.9 26.5 15

6 5

14 9

17 7 9

10 27 8

4.1 3.4

9.5 6.2

11.6 4.8 6.1

6.8 18.3 5.4

147 147

147 147 147 147 147

147 147 147

100 100

100 100 100 100 100

100 100 100


32

Table 11B Motives for varying credit terms offered by type of firms : the small enterprises


N % [2]

N % [3]

N % [4]

N % [5]

N % Total



Competitive motives






93

117

64 80 40 31 66

73 30 45

45.6 57.3

31.4 39.2 19.6 15.2 32.3

35.8 14.7 22.1

57 33

29 54 27 38 58

43 19 27

27.9 16.2

14.2 26.5 13.2 18.6 28.4

21.1 9.3

13.2

32 32

60 42 68 77 43

35 66 72

15.7 15.7

29.4 20.6 33.3 37.7 21.1

17.2 32.3 35.3

14 12

28 15 41 41 23

29 57 44

6.9 5.9

13.7 7.3

20.2 20.1 11.3

14.2 27.9 21.6

8

10

23 13 28 17 14

24 32 16

3.9 4.9

11.3 6.4

13.7 8.4 6.9

11.7 15.8 7.8

204 204

204 204 204 204 204

204 204 204

100 100

100 100 100 100 100

100 100 100


Table 11C Motives for varying credit terms offered by type of firms : the medium enterprises


N % [2]

N % [3]

N % [4]

N % [5]

N % Total



Competitive motives






80 92

50 60 32 24 41

47 27 44

49.4 56.8

30.9 37

19.7 14.8 25.3

29 16.7 27.2

40 25

19 33 24 27 44

42 23 25

24.7 15.4

11.7 20.4 14.8 16.7 27.2

25.9 14.2 15.4

25 24

48 37 51 65 37

25 48 57

15.4 14.8

29.6 22.8 31.5 40.1 22.8

15.4 29.6 35.2

10 14

30 21 35 32 24

23 42 21

6.2 8.7

18.5 13

21.6 19.7 14.8

14.2 25.9 13

7 7

15 11 20 14 16

25 22 15

4.3 4.3

9.3 6.8

12.4 8.7 9.9

15.4 13.6 9.2

162 162

162 162 162 162 162

162 162 162

100 100

100 100 100 100 100

100 100 100


33

Table 12

Credit periods

Credit periods (days) N % firms Cash [1-7]

[8-14] [15-29]

[30] [31-46] [47-60] [60 et +]

28 21 10 25 47 71

135 176

5.4 4.2 1.9 4.8 9.2

13.8 26.3 34.4

Total 513 100

Table 13 Credit periods by type of market

Credit periods (days) Local

N % Regional

N % National

N % International

N % Cash [1-7]

[8-14] [15-29]

[30] [31-46] [47-60] [60 et +]

12 8 3 6

14 9 9

13

16.2 10.8

4 8.1

18.9 12.2 12.2 17.6

16 10 2

11 16 42 67 66

6.9 4.3 0.9 4.8 6.9

18.3 29.1 28.8

0 2 4 7 9

12 43 59

0 1.5 3

5.1 6.6 8.9

31.6 43.3

0 1 1 1 8 8

16 38

0 1.4 1.4 1.4

10.9 10.9 21.9 52.1

Total 74 100 230 100 136 100 73 100

34

Tableau 14

Credit periods by type of firms

Credit periods (days) Micro-enterprises N %

Small enterprises N %

Medium enterprises N %

Cash [1-7]

[8-14] [15-29]

[30] [31-46] [47-60] [60 et +]

15 10 3

10 18 23 33 35

10.2 6.8 2.1 6.8

12.2 15.6 22.5 23.8

10 8 4 8

17 29 65 63

4.9 3.9 2

3.9 8.3

14.2 31.9 30.9

3 3 3 7

12 19 37 78

1.9 1.9 1.9 4.3 7.4

11.7 22.8 48.1

Total 147 100 204 100 162 100

Table 15 Use of discounts with two-part terms in credit offered

Frequency of use N %

Never Seldom

Occasionally Frequently

Always

253 129 80 28 23

49.3 25.1 15.6 5.5 4.5

Total 513 100

35

Table 16 Use of discounts with two-part terms in credit offered by industry classification

(in % of the number of firms)

Frequency of use EA %

EB %

EC %

ED %

EE %

EF %

EH %

EJ %

EK %

EN %

EP %

EQ %

Never Seldom


Always

25 50 25 0 0

31.2 50

18.8 0 0

21.6 29.7 24.3 8.2

16.2

62.5 37.5

0 0 0

46.9 21.9 31.2

0 0

37.4 30.8 18.7 5.5 7.6

59.7 19.4 11.9

9 0

48.6 23.2 15.7

7 5.5

71.4 14.3 10.7 3.6 0

78.9 21.1

0 0 0

66.7 33.3

0 0 0

100 0 0 0 0

Total 100 100 100 100 100 100 100 100 100 100 100 100

Table 17 Use of discounts with two-part terms in credit offered by type of market

Frequency of use Local

N % Regional

N % National

N % International

N % Never

Seldom Occasionally Frequently

Always

53 14 6 0 1

71.6 18.9 8.2 0

1.3

118 64 26 13 9

51.3 27.8 11.3 5.7 3.9

51 35 34 9 7

37.5 25.7 25 6.6 5.2

31 16 14 6 6

42.5 21.9 19.2 8.2 8.2

Total 74 100 230 100 136 100 73 100

Table 18 Frequency of late payments in credit collection

Frequency of late payments N %

Never Seldom


Always

21 99 214 167 12

4.1 19.3 41.7 32.6 2.3

Total 513 100

Table 19 Variable definitions and summary statistics of the model of trade credit supply in 2002

(mean, median, standard deviation and number of observations) Type of variables Name Definition of variables Mean Median Std. dev. Obs.

Dependent variables

TCI TCD

NT vs CT TPT vs CT TPT vs NT

Accounts receivable / turnover (Accounts receivable * 360) / turnover = 1 if offering net terms, 0 if offering cash terms = 1 if offering two-part terms, 0 if offering cash terms = 1 if offering two-part terms, 0 if offering net terms

0.18 64.29

0.94 0.63 0.10

0.19 63

1 1 0

0.15 44.65

0.21 0.46 0.30

513 513

465 76 485

Supplier’s characteristics

LTA

LOC REG NAT INT

AGE AGE²

PROF

RAT

Logarithm of total assets value = 1 if the firm sale’s market is local, 0 otherwise = 1 if the firm sale’s market is regional, 0 otherwise = 1 if the firm sale’s market is national, 0 otherwise = 1 if the firm sale’s market is international, 0 otherwise Logarithm of (1 + firm’s age) Logarithm of (1 + firm’s age)² Gross margin of exploitation / turnover = 1 if short-term bank credit is rationed, 0 otherwise

12.65

0.14 0.45 0.27 0.14

2.58 5.17

0.08

0.10

12.17

0 0 0 0

2.38 4.75

0.06

0

11.27

0.35 0.44 0.50 0.35

0.74 1.31

0.17

0.29

513

513 513 513 513

513 513

513

513

Products/services characteristics

QUA FCP INN

COM

= 1 if the quality of products/services sold is not easily to inspect, 0 otherwise = 1 if products/services sold are fast changing, 0 otherwise = 1 if products/services sold are innovating, 0 otherwise

= 1 if the firm sale’s market is strongly competitive, 0 otherwise

0.06 0.22 0.29

0.88

0 0 0

1

0.23 0.41 0.45

0.33

513 513 513

513

Independent variables

Demand characteristics

NAC SCN

RET WHL PRO

LOW EQU HIG

SEA

Logarithm of firm’s number of active customers Turnover / number of active customers = 1 if firm’s customer base is mainly composed of retailers, 0 otherwise = 1 if firm’s customer base is mainly composed of wholesalers, 0 otherwise = 1 if firm’s customer base is mainly composed of producers, 0 otherwise = 1 if firm’s customer base is mainly composed of lower-sized firms, 0 otherwise = 1 if firm’s customer base is mainly composed of equal-sized firms, 0 otherwise = 1 if firm’s customer base is mainly composed of higher-sized firms, 0 otherwise = 1 the firm faces a seasonal demand, 0 otherwise

6.41 190.73

0.37 0.22 0.41

0.35 0.22 0.43

0.48

5.7 11.02

0 0 0

0 0 0

0

7.21 1818

0.48 0.42 0.49

0.48 0.41 0.49

0.49

513 513

513 513 513

513 513 513

513

36

Table 20 Estimates of trade credit supply

Independent variables are relative to supplier’s characteristics [1], products/services sold [2] and demand [3]. Models I and II : determinants of the variable (TCI), O.L.S estimates, standard deviations in brackets ; III and IV : determinants of the variable (TCD), ordered probit estimates, khi-square in brackets ; V, VI and VII : determinants of variables {NT(1) vs CT(0)}, {TPT(1) vs CT(0)} and {TPT(1) vs NT(0)}, logit estimates, khi-square in brackets. 1,2 and 3 : significant at 1%, 5% et 10%, respectively.

Variables I II III IV V VI VII [1]

LTA

REG

NAT

INT

AGE

AGE²

PROF

RAT

.0121

(.027) - - -

.0172

(.008) -.0052

(.002) .0512

(.026) -.0101

(.003)

-

.007 (.015) .0223

(.012) .0501

(.018) .0142

(.006) -.0042

(.002) .0732

(.035) -.0122

(.005)

.3541

(7.57) - - -

.0412

(5.21) -.0042

(4.37) .2181

(6.80) -.6972

(4.48)

-

.132 (0.87) .2733

(2.89) .6471

(7.28) .0252

(5.53) -.0022

(4.57) .3151

(7.46) -.3622

(5.32)

-

.879 (1.04) 1.01

(1.16) .9783

(2.83) .115

(1.57) -

.0323

(2.85) -1.37 (2.06)

-

.942 (.891) .876

(1.22) 1.171

(7.86) .8263

(3.77) -

-.734 (.573) -.779 (1.92)

-

.706 (1.54) .7953

(3.26) 1.751

(7.47) .1101

(8.51) -

-.021 (.292) .043

(.571) [2]

QUA

FCP

INN

COM

.0332

(.015) -.0222

(.010) .004

(.010) .032

(.027)

.0392

(.016) -.0282

(.012) .006

(.010) .027

(.033)

.6473

(2.86) -.1342

(4.25) .048 (.59) .533

(1.15)

.4652

(5.32) -.1943

(2.88) .075

(.780) .282

(1.96)

.170

(.451) -.7442

(5.85) 2.402

(4.74) 1.16

(1.81)

.762

(.803) -1.432

(5.07) 1.063

(3.81) 1.482

(4.94)

-.1103

(3.67) -.993 (2.21) .037

(.532) .6343

(2.97) [3]

NAC

SCN

WHL

PRO

EQU

HIG

SEA

-.0231

(.003) -

.015 (.009) .0451

(.012) .018

(.012) .0253

(.022) -.010 (.009)

-

.0312

(.014) .014

(.001) .0471

(.013) .017

(.013) .022

(.017) -.008 (.009)

-.1451

(7.79) -

.208 (.894) .6531

(9.03) .6613

(1.88) .761

(1.72) -.219 (1.29)

-

.0362

(4.98) .074

(.515) .3941

(8.14) .356

(1.92) .380

(1.91) -.110 (1.13)

-.113 (2.35)

-

.119 (.445) .9221

(6.99) 1.74

(2.46) 2.363

(2.83) .398

(1.01)

-.757 (.658)

-

1.33 (.648) 1.833

(3.33) .877

(1.08) 1.06

(.971) .144

(1.59)

-.335 (1.36)

-

.232 (.486) -.6812

(4.72) .053

(.494) .092

(1.83) .029

(1.67)

N R²

F-test -2LL

513 0.22 8.211

-

513 0.21 7.521

-

513

- -

561.34

513

- -

559.48

465

- -

425.32

76 - -

174.32

485

- -

442.31

37

Table 21 Variable definitions and summary statistics of the model of trade credit demand in 2002

(mean, median, standard deviation and number of observations)

Type of variables Name Definition of variables Mean Median Std. dev. Obs.

Dependent variables

AP

LTP

Accounts payable / total assets = 1 if late payments to suppliers are frequent or systematic = 0 otherwise

0.27

0.17

0.26

0

0.16

0.39

513

513

Transaction component

STO

STUR NAS

TNV

Stocks / total assets (Stocks * 360) / turnover Logarithm of firm’s number of active suppliers Logarithm of turnover

0.20

48.66 4.99

12.74

0.16

34.60 4.40

12.65

0.17

50.74 5.59

11.62

513

513 513

513

Independent variables

Financing component

SCOR LEV

AGE RAM

TPT

STD

RAT

Score variable Conan-Holder Total debt / equity Logarithm of (1 + firm’s age) = 1 if firm’s form of business is proprietorship or partnership = 0 otherwise = 1 if discounts with two-part terms are offered by suppliers = 0 otherwise Short-term bank credit / total assets = 1 if short-term bank credit is rationed = 0 otherwise

17.46 2.21

2.58 0.26

0.13

0.09

0.10

16.16 2.02

2.38

0

0

0.07

0

10.76 1.15

0.74 0.44

0.33

0.11

0.29

513 513

513 513

513

513

513

38

Table 22 Estimates of trade credit demand

Independent variables are relative to the transaction component [1] and the financing component [2] of trade credit demand. Models I and II : determinants of the variable (AP), O.L.S estimates, standard deviations in brackets ; III and IV : determinants of the variable (LTP), logit estimates, khi-square in brackets. 1,2 and 3 : significant at 1%, 5% et 10%, respectively.

Variables I II III III

[1]

STO

STUR

NAS

TNV

.0581

(.021) -.0011

(.000) .0222

(.008) .0431

(.005)

.0531

(.020) -.0021

(.000) .0252

(.009) .0411

(.005)

-.241 (.823) .016

(.595) .382

(1.24) .107

(.958)

-197 (.849) .019

(1.04) .419 (.86) .141

(.526) [2]

SCOR

LEV

AGE

RAM

TPT

STD

RAT

-.0062

(.003) .0241

(.004) -.0032

(.001) .027

(.023) -.006 (.005) .0481

(.010) -

-.0072

(.003) .0251

(.004) -.002 3

(.001) .030

(.264) -.007 (.006) .0461

(.010) .0581

(.021)

-.0511

(7.34) .3092

(5.21) -.0152

(4.32) -

-.1252

(3.21) .6493

(3.21) -

-.0582

(5.87) .3252

(6.12) -.0123

(3.63) .439

(1.06) -.157 3

(3.48) .511

(2.12) .1091

(8.35)

Intercept

N R²

F-test -2 LL

-.6051

(.142)

513 0.38

15.781

-

-.7311

(.127)

513 0.44

16.681

-

-1.831

(7.32)

513 - -

521.36

-1.971

(8.08)

513 - -

546.62

39

Appendix 1 Algebraic manipulations of the model of trade credit demand

Elliehausen and Wolken [1993] Presented equations are the following : [1] DTC* = TC + FC [2] TC = F (PUR, ∆DEL, RLA, CCLA) TC = a0 + a1 PUR + a2 ∆DEL + a3 RLA + a4 CCLA [3] FC = F (FR, BR, PTC, PBC) FC = b0 + b1 FR + b2 BR + b3 PTC + b4 PBC [4] DBC = F (PBC) DBC = c0 + c1 PBC PBC = (DBC - c0) / c1

[5] DTC = AP

[6] DTC*- DTC = F (FR, BR)

DTC*- DTC = d1 FR + d2 BR

[7] DTC* = AP + F (FR, BR)

DTC* = AP + d1 FR + d2 BR

Introducing equations [2] and [3] in equation [1] yields :

[1] DTC* = [a0 + a1 PUR + a2 ∆DEL + a3 RLA + a4 CCLA]

+ [b0 + b1 FR + b2 BR + b3 PTC + b4 PBC]

Combining equations [4] and [7] with equation [1] is :

DTC* = TC + FC = AP + d1 FR + d2 BR

AP = DTC* - d1 FR - d2 BR

AP = [a0 + a1 PUR + a2 ∆DEL + a3 RLA + a4 CCLA]

+ {b0 + b1 FR + b2 BR + b3 PTC + b4 [(DBC - c0) / c1] }

- d1 FR - d2 BR

Following [5], after simplifications :

DTC = AP = [a0 + b0 - b4 (c0 / c1)]

+ [a1 PUR + a2 ∆DEL + a3 RLA + a4 CCLA]

+ {(b1 - d1) FR + (b2 - d2) BR + b3 PTC + [(b4 / c1) DBC] } + ε

Then : DTC = AP = α + TC + FC + ε

40

Appendix 2 Main characteristics of the Commission Recommendation of 3 April 1996

concerning the definition of small and medium-sized enterprises

Small and medium-sized enterprises, hereinafter referred to as “SMEs”, are defined as enterprises which :

- have fewer than 250 employees - and have either : an annual turnover not exceeding ECU 40 million, or an annual balance-sheet total not exceeding ECU 27 million, - conform to the criterion of independence as defined below.

Where it is necessary to distinguish between small and medium-sized enterprises, the “small enterprise” is defined as an enterprise which :

- has fewer than 50 employees - and has either : an annual turnover not exceeding ECU 7 million, or an annual balance-sheet total not exceeding ECU 5 million, - conforms to the criterion of independence as defined below.

Independent enterprises are those which are not owned as to 25 % or more of the capital or the

voting rights by one enterprise, or jointly by several enterprises, falling outside the definition of an SME or a small enterprise, whichever may apply. This threshold may be exceeded in the following two cases : - if the enterprise is held by public investment corporations, venture capital companies or institutional investors, provided no control is exercised either individually or jointly, - if the capital is spread in such a way that it is not possible to determine by whom it is held and if the enterprise declares that it can legitimately presume that it is not owned as to 25 % or more by one enterprise, or jointly by several enterprises, falling outside the definitions of an SME or a small enterprise, whichever may apply. In calculating the thresholds referred above, it is therefore necessary to cumulate the relevant

figures for the beneficiary enterprise and for all the enterprises which it directly or indirectly controls through possession of 25 % or more of the capital or of the voting rights.

Where it is necessary to distinguish micro-enterprises from other SMEs, these are defined as

enterprises having fewer than 10 employees.

41

References

BARDES B. [2002], “Délais de Paiement et Solde du Crédit Interentreprises de 1989 à 2001”, Bulletin de la Banque de France, 108, p.63-79.

BARDES B., COCHONNEAU D., NIVELAIS M. [2002], “Délais de Paiement et Solde du Crédit Interentreprises: Statistiques de 1992 à 2001 en Nomenclature NAF”, Collection Entreprises, Observatoire des entreprises, Banque de France, 2002.

BAROMETRE EUROFACTOR [2002], http://www.eurofactor.com/gfr/pdf/BAROMETRE_EUROFACTOR_2002.pdf

BAUM C.F., CAGLAYAN M., OZKAN N. [2003], “The Impact of Macroeconomic Uncertainty on Trade Credit for Non-Financial Firms”, Working Paper 566, Boston College.

BERNANKE B.S., BLINDER A. [1988], “Credit, Money and Aggregate Demand”, American Economic Review, 78, p. 435-439.

BIAIS B., GOLLIER C. [1997], “Trade Credit and Credit Rationing”, Review of Financial Studies, 10, p. 903-937.

BRECHLING F.P.R., LIPSEY R.G. [1963], “Trade Credit and Monetary Policy”, Economic Journal, 73, p.618-641.

BRENNAN M., MAKSIMOVIC V., ZECHNER J. [1988], “Vendor Financing”, Journal of Finance, 43, p.1127-1141.

CARLING K., LUNDBERG S. [2002], “Bank lending, Geographical distance, and Credit Risk: An Empirical Assessment of the Church Tower Principle”, Riksbankens Working paper series 144, Stockholm.

CHANT E.M., WALKER D.A. [1988], “Small Business Demand for Trade Credit,” Applied Economics, 20, p.861-876.

CONAN D., HOLDER M. [1979], “Variables explicatives de performance et contrôle de gestion dans les P.M.I.", Thèse d'Etat, CERG, Université Paris Dauphine.

CUEVAS C., FAFCHAMPS M., HANSON R., MOLL P., SRIVASTAVA P. [1993], “Case Studies of Enterprise Finance in Ghana, RPED Case Study Series, Washington DC, World Bank.

CUNAT V. [2001], “Trade credit: suppliers as debt collectors and insurance providers”, Mimeo, Financial Markets Group, London School of Economics.

DANIELSON M.G., SCOTT J.A. [2000], “Additional Evidence on the Use of Trade Credit by Small Firms: the Role of Trade Credit Discount”, Manuscript.

DIETSCH M. [1990], “Le Crédit Interentreprises : Coûts et Avantages”, Economie et Statistique, 236, p.65-79. DIETSCH M. [1998], “Atouts et Handicaps du Crédit Client face au Crédit Bancaire”, Revue d'Economie

Financière, 46, p.175-193. DIETSCH M., KREMP E. [1998], “Le Crédit Interentreprises bénéficie plus aux Grandes Entreprises qu'aux

PME”, Economie et Statistique, 314, p.25-37. DRAULT T. [2001], “Lutte contre les Retard de Paiement des Transactions”, http://www.net-

iris.com/publication/author/document.php3?document=32. DUN & BRADSTREET [2001/2002], http://dbfrance.dnb.com/French/default.htm ELLIEHAUSEN G.E., Wolken J.D. [1993], “The Demand for Trade Credit: An Investigation of Motives of

Trade Credit Use by Small Businesses”, Working paper, Board of Governors, Federal Reserve System. EMERY G.W. [1984], “A Pure Financial Explanation for Trade Credit”, Journal of Financial and Quantitative

Analysis, 19, p.271-285. EMERY G.W. [1987], “An Optimal Financial Response to Variable Demand”, Journal

of Financial and Quantitative Analysis, 22, p. 209-225. FAFCHAMPS M. [1997], “Trade Credit in Zimbabwean Manufacturing”, World Development, 25, p.795-815. FAFCHAMPS M. [2000], “Ethnicity and Credit in African Manufacturing”, Journal of Development Economics,

61, p.205-235. FAFCHAMPS M., PENDER J., ROBINSON E. [1995], “Enterprise Finance in Zimbabwe”, RPED Case Study

Series, Washington DC, World Bank. FERRIS J.S. [1981], “A Transactions Theory of Trade Credit Use”, Quarterly Journal of Economics, 96, p. 243-

270. HOL S., WIJST N.V.D [2002], “Trade Credit in Europe”, 31st Meeting of the European Working Group on

Financial Modeling (EWGFM), November 7-9, Agia Napa, Cyprus. ISAKSSON A. [2002], “Trade Credit in Kenyan Manufacturing: Evidence from plant-level

data”, Working Paper, Statistical and Information Networks (UNIDO). JAFFEE D.M. [1971], “Credit Rationing and the Commercial Loan Market”, New York: Wiley. JAIN N. [2001], “Monitoring Costs and Trade Credit”, Quarterly Review of Economics and Finance, 41, p.89-

110.

42

JOHNSON S., McMILLAN J., WOODRUFF C. [2002], “Property Rights and Finance”, Working Paper No. 8852, Cambridge, NBER.

KNEESHAW J.T. [1995], “A Survey of Non-financial Sector Balance Sheets in Industrialized Countries: Implications for the Monetary Policy Transmission Mechanism”, BIS Working Paper No. 25, Basle: Bank for International Settlements.

KOHLER M., BRITTON E., YATES T. [2000], “Trade Credit and the Monetary Transmission Mechanism”, Working Paper ISSN 1368-5562, Bank of England.

LAFFER A.B. [1970], “Trade Credit and the Money Market”, Journal of Political Economy, 78, p.239-267. LEE Y.W., STOWE J.D. [1993], “Product Risk, Asymmetric Information, and Trade Credit”, Journal of

Financial and Quantitative Analysis, 28, p.285-300. LONG M.S., MALITZ I.B., RAVID S.A. [1993], “Trade Credit, Quality Guarantees, and Product

Marketability”, Financial Management, 22, p.117-127. MAROTTA G. [2001], “Is Trade Credit More expensive than Bank Loans? Evidence from Italian Firm-Level

Data”, Materiali di discussione, Dipartimento di Economia Politica, University of Modena and Reggio Emilia.

MATEUT S., BOUGHEAS S., MIZEN P. [2003], “Trade Credit, Bank Lending and Monetary Policy Transmission”, Economics Working Papers ECO2003/02, European University Institute.

MELTZER A.H. [1960], “Mercantile Credit, Monetary Policy, and Size of Firms”, Review of Economics and Statistics, 42, p.429-437.

MIAN S.L., SMITH C.W. [1992], “Accounts Receivable Management Policy: Theory and Evidence”, Journal of Finance, 47, p.169-200.

MYERS S. [1984], “The Capital Structure Puzzle”, Journal of Finance, 39, p.575-592. NADIRI M.I. [1969], “The Determinants of Trade Credit in the U.S. Total Manufacturing Sector”,

Econometrica, 37, p.408-423. NG C.K., SMITH J.K, SMITH R.L, [1999], “Evidence on the Determinants of Credit Terms Used in Interfirm

Trade”, Journal of Finance, 54, p.1109-1129. NILSEN J.H. [1999], “Trade Credit and the Bank Lending Channel”, Journal of Money, Credit and Banking, 34,

p.226-253. PETERSEN M.A., RAJAN R.G. [1994], “The Benefits of Lending Relationships: Evidence from Small

Business Data”, Journal of Finance, 49, p.3-37. PETERSEN M.A., RAJAN R.G. [1997], “Trade Credit: Theories and Evidence”, Review of Financial Studies,

10, p.661-691. SCHWARTZ R.A. [1974], “An Economic Model of Trade Credit”, Journal of Financial and Quantitative

Analysis, 9, p.643-657. SCHWARTZ R.A., WHITCOMB D.K [1978], “Implicit Transfers in the Extension of Trade Credit”, in

Redistribution Through The Financial System: The Grants Economics of Money and Credit, Boulding, . E. and T.F. Wilson (eds), Praeger Special Studies, New York, p.191-208.

SINGLETON C., SUMMERS B., WILSON N. [1999], “Small Business Demand for Trade Credit, Credit rationing and the Late Payment of Commercial Debt: An Empirical Study”, in Management Buy-Outs and Venture Capital: Into the Next Millenium, M. Wright and K.Robbie (eds), Edward Elgar Publishing, p.181-201.

SMITH J.K. [1980], Readings on the Management of Working Capital, West Publishing, St. Paul, Minnesota. SMITH J.K. [1987], “Trade Credit and Informational Asymmetry”, Journal of Finance, 42, p.863-872. STIGLITZ J.E., WEISS A. [1981], “Credit Rationing in Markets with Imperfect Information”, American

Economic Review, 71, p.393–410. SUMMERS B., WILSON N. [2002a], “Trade Credit Terms Offered by Small Firms: Survey Evidence and

Empirical Analysis”, Journal of Business Finance and Accounting, 29, p.317-351. SUMMERS B., WILSON N. [2002b], “An Empirical Investigation of Trade Credit Use: A Note”, International

Journal of the Economics of Business, 9, p.252-270. SUMMERS B., WILSON N. [2003], “Trade Credit and Customer Relationships”, Managerial and Decision

Economics, 24, p.439-455. UFB LOCABAIL [1999], http://www.bnpparibas-leasegroup.com/enquetes/pme_pmi_id11_p.htm WILNER B.S. [2000], “The Exploitation of Relationships in Financial Distress: The Case of Trade Credit”,

Journal of Finance, 55, p.153-178.

43

trade credit, linked to a temporal gap between delivery and

Documents

trade credit small

practices of trade credit

use of trade credit

role of trade credit

introductiontrade credit

trade debts

bank credit rationing

interfirm credit